Lunit INSIGHT DBT is a computer-assisted detection and diagnosis (CADe/x) software intended to be used concurrently by interpreting physicians to aid in the detection and characterization of suspected lesions for breast cancer in digital breast tomosynthesis (DBT) exams from compatible DBT systems. Through the analysis. the regions of soft tissue lesions and calcifications are marked with an abnormality score indicating the likelihood of the presence of malignancy for each lesion. Lunit INSIGHT DBT uses screening mammograms of the female population.

Lunit INSIGHT DBT is not intended as a replacement for a complete interpreting physician's review or their clinical judgment that takes into account other relevant information from the image or patient history.

Device Description

Lunit INSIGHT DBT is a computer-assisted detection/diagnosis (CADe/x) software as a medical device that provides information about the presence, location and characteristics of lesions suspicious for breast cancer to assist interpreting physicians in making diagnostic decisions when reading digital breast tomosynthesis (DBT) images. The software automatically analyzes digital breast tomosynthesis slices via artificial intelligence technology that has been trained via deep learning.

For each DBT case, Lunit INSIGHT DBT generates an artificial intelligence analysis results that include the lesion type, location, lesion-level case-level score, and outline of the regions suspected of breast cancer. This peripheral information intends to augment the physician's workflow to better aid in detection and diagnosis of breast cancer.

AI/ML Overview

Here's a breakdown of the acceptance criteria and the study proving the device, Lunit INSIGHT DBT, meets them, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

Performance Metric	Acceptance Criteria	Reported Device Performance	Statistical Significance / Comment
Standalone Performance
AUROC	> 0.903 (mean AUROC of predicate device K211678)	0.928 (95% Cl: 0.917 - 0.939)	p < 0.0001 (exceeded criteria)
Clinical Assessment (MRMC Study with CAD Assistance)
Patient-level LOS AUROC	CAD-assisted performance superior to CAD-unassisted performance with statistical significance	CAD-unassisted AUROC: 0.897 (95% Cl 0.858 - 0.936)CAD-assisted AUROC: 0.915 (95% Cl: 0.874 - 0.955)	Inter-test difference: 0.017 (95% Cl: 0.000 - 0.034, P = 0.0498) - met criteria.

2. Sample Size Used for the Test Set and Data Provenance

Standalone Performance Test Set: 2,202 DBT exams (1,100 negative/benign, 1,102 cancer cases).
- Data Provenance: Collected at multiple imaging facilities in the US. The data was collected consecutively.
Clinical Assessment (MRMC) Test Set: 258 DBT exams (65 cancer cases, 193 non-cancer cases, comprising 128 normal and 65 benign cases).
- Data Provenance: Acquired from US clinical centers.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Their Qualifications

Standalone Performance Test Set: The text states ground truth localization was "derived based on the radiologic review and annotation by multiple MQSA qualified ground truthers." The exact number of experts is not specified.
- Qualifications: "MQSA qualified ground truthers."
Clinical Assessment (MRMC) Test Set: The ground truth for the cases used in the MRMC study is implicitly established by the case classification (cancer vs. non-cancer). It's not explicitly stated how many experts established the underlying ground truth for these 258 cases, beyond the radiologists participating in the MRMC study itself. The readers for the MRMC study were "a total of 15 MQSA qualified and US board-certified radiologists."

4. Adjudication Method for the Test Set

Standalone Performance Test Set: Ground truth was established through "binary classification of each case based on clinical supporting data, particularly pathology reports for cancer and biopsy-proven benign cases, followed by localization which was derived based on the radiologic review and annotation by multiple MQSA qualified ground truthers." This suggests an expert consensus/review process for localization, likely involving reconciliation or multiple reads. The specific type of adjudication (e.g., 2+1, 3+1) is not explicitly detailed.
Clinical Assessment (MRMC) Test Set: The ground truth for the MRMC study seems to rely on the pre-established classification of cases (cancer/non-cancer) based on clinical and pathological data. The radiologists in the MRMC study were evaluating cases against this existing ground truth, not establishing it in an adjudicated reading session.

5. Was a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Done? If so, what was the effect size of how much human readers improve with AI vs. without AI assistance?

Yes, an MRMC comparative effectiveness study was done.
Effect Size of Improvement:
- CAD-unassisted AUROC: 0.897
- CAD-assisted AUROC: 0.915
- Inter-test difference (effect size): 0.017 (Absolute difference in AUROC). This indicates an improvement in AUROC of 0.017 when radiologists were assisted by Lunit INSIGHT DBT. The p-value of 0.0498 indicates this improvement was statistically significant.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done

Yes, a standalone performance study was done.
- Performance: AUROC of 0.928 (95% Cl: 0.917 - 0.939).

7. The Type of Ground Truth Used

Standalone Performance and Clinical Assessment Test Sets: The ground truth was established primarily through clinical supporting data, specifically pathology reports for cancer and biopsy-proven benign cases, further supplemented by radiologic review and annotation by MQSA qualified ground truthers for localization. This can be categorized as a combination of pathology and expert consensus/review.

8. The Sample Size for the Training Set

The document states that the "dataset used in the standalone performance test was independent from the dataset used for development of the artificial intelligence algorithm." However, it does not specify the sample size of the training set used for the development of Lunit INSIGHT DBT.

9. How the Ground Truth for the Training Set Was Established

The document mentions that the training dataset was separate from the test dataset. While it details how the ground truth was established for the test set (pathology, biopsy, radiologic review/annotation), it does not explicitly describe how the ground truth was established for the training set. Being a deep learning model, it's highly probable the training data also relied on verified diagnoses, likely similarly derived from pathology/biopsy given the nature of CADe/x devices for breast cancer.

Summary

{0}------------------------------------------------

November 6, 2023

Image /page/0/Picture/1 description: The image shows the logo of the U.S. Food and Drug Administration (FDA). On the left is the Department of Health & Human Services logo. To the right of that is the FDA logo, which consists of the letters "FDA" in a blue square, followed by the words "U.S. FOOD & DRUG" in blue, with the word "ADMINISTRATION" underneath.

Lunit Inc. % Hyung Tak Han Regulatory Affairs Specialist 4-8 F. 374 Gangnam-daero, Gangnam-gu SEOUL. 06241 SOUTH KOREA

Re: K231470

Trade/Device Name: Lunit INSIGHT DBT Regulation Number: 21 CFR 892.2090 Regulation Name: Radiological Computer Assisted Detection And Diagnosis Software Regulatory Class: Class II Product Code: QDQ Dated: October 4, 2023 Received: October 4, 2023

Dear Hyung Tak Han:

We have reviewed your section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (the Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. Although this letter refers to your product as a device, please be aware that some cleared products may instead be combination products. The 510(k) Premarket Notification Database available at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm identifies combination product submissions. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading.

If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.

Additional information about changes that may require a new premarket notification are provided in the FDA guidance documents entitled "Deciding When to Submit a 510(k) for a Change to an Existing Device" (https://www.fda.gov/media/99812/download) and "Deciding When to Submit a 510(k) for a Software Change to an Existing Device" (https://www.fda.gov/media/99785/download).

{1}------------------------------------------------

Your device is also subject to, among other requirements, the Quality System (QS) regulation (21 CFR Part 820), which includes, but is not limited to, 21 CFR 820.30, Design controls; 21 CFR 820.90, Nonconforming product; and 21 CFR 820.100, Corrective and preventive action. Please note that regardless of whether a change requires premarket review, the QS regulation requires device manufacturers to review and approve changes to device design and production (21 CFR 820.30 and 21 CFR 820.70) and document changes and approvals in the device master record (21 CFR 820.181).

Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting of medical device-related adverse events) (21 CFR Part 803) for devices or postmarketing safety reporting (21 CFR Part 4, Subpart B) for combination products (see https://www.fda.gov/combination-products/guidance-regulatory-information/postmarketing-safety-reportingcombination-products); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820) for devices or current good manufacturing practices (21 CFR Part 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR Parts 1000-1050.

Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to https://www.fda.gov/medical-device-safety/medical-device-reportingmdr-how-report-medical-device-problems.

For comprehensive regulatory information about mediation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/medicaldevices/device-advice-comprehensive-regulatory-assistance) and CDRH Learn (https://www.fda.gov/training-and-continuing-education/cdrh-learn). Additionally, you may contact the Division of Industry and Consumer Education (DICE) to ask a question about a specific regulatory topic. See the DICE website (https://www.fda.gov/medical-device-advice-comprehensive-regulatoryassistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).

Sincerely,

Yanna S. Kang -S

Yanna Kang, Ph.D. Assistant Director Mammography and Ultrasound Team DHT8C: Division of Radiological Imaging and Radiation Therapy Devices OHT8: Office of Radiological Health Office of Product Evaluation and Quality Center for Devices and Radiological Health

{2}------------------------------------------------

Indications for Use

Submission Number (if known)

K231470

Device Name

Lunit INSIGHT DBT

Indications for Use (Describe)

Type of Use (Select one or both, as applicable)

Prescription Use (Part 21 CFR 801 Subpart D)

Over-The-Counter Use (21 CFR 801 Subpart C)

CONTINUE ON A SEPARATE PAGE IF NEEDED.

This section applies only to requirements of the Paperwork Reduction Act of 1995.

DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW.

The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:

Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff(@fda.hhs.gov

"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."

{3}------------------------------------------------

Image /page/3/Picture/0 description: The image contains the logo for Lunit, a medical AI company. The logo consists of a blue circular icon with a white molecular-like structure inside, followed by the company name "Lunit" in bold, black font. A registered trademark symbol is placed next to the name.

Lunit Inc. 4-8 F, 374, Gangnam-daero, Gangnam-gu, Seoul, 06241, Republic of Korea www.lunit.io

Page 1/6

510(k) Summary

Lunit INSIGHT DBT (K231470)

This 510(k) summary of safety and effectiveness information is prepared in accordance with the requirements of 21 CFR §807.92.

1. Submitter

Applicant (Manufacturer)	Lunit Inc.4-8 F, 374, Gangnam-daero, Gangnam-gu,Seoul, 06241, Republic of KoreaTel: + 82-70-5066-0849FAX: +82-2-6919-2702E-mail: ra_rad@lunit.io
Primary Correspondent	Harry Hyung Tak HanRegulatory Affairs SpecialistEmail: hhan@lunit.io
Secondary Correspondent	Suhyoung BahkRegulatory Affairs SpecialistEmail: sbahk@lunit.io
Date Prepared	2023. 11. 03

Device Names and Classifications 2.

Subject Device

Name of Device	Lunit INSIGHT DBT
Classification Name	Radiological Computer Assisted Detection/Diagnosis Software For SuspiciousLesions For Cancer
Regulation	21 CFR 892.2090
Regulatory Class	Class II
Product Code	QDQ

{4}------------------------------------------------

Image /page/4/Picture/0 description: The image shows the Lunit logo. The logo consists of a blue circle with a white molecular-like structure inside, followed by the word "Lunit" in black, with a registered trademark symbol next to it. The logo is simple and modern, and the colors are eye-catching.

nam-daero. Gangnam-gu. 06241. Republic of Korea

Page 2/6

Predicate Device

Name of Device	Lunit INSIGHT MMG
Classification Name	Radiological Computer Assisted Detection/Diagnosis Software For SuspiciousLesions For Cancer
Regulation	21 CFR 892.2090
Regulatory Class	Class II
Product Code	QDQ
Submission Number	K211678

3. Device Description

4. Indication for Use

Lunit INSIGHT DBT is a computer-assisted detection and diagnosis (CADe/x) software intended to be used concurrently by interpreting physicians to aid in the detection and characterization of suspected lesions for breast cancer in digital breast tomosynthesis (DBT) exams from compatible DBT systems. Through the analysis, the regions of soft tissue lesions and calcifications are marked with an abnormality score indicating the likelihood of the presence of malignancy for each lesion. Lunit INSIGHT DBT uses screening mammograms of the female population.

{5}------------------------------------------------

Image /page/5/Picture/0 description: The image contains the logo for Lunit. The logo consists of a blue circle with a white molecular-like structure inside, followed by the word "Lunit" in black, with a registered trademark symbol next to it. The logo is simple and modern, with a focus on the company's name.

Lunit Inc.
4-8 F, 374, Gangnam-daero, Gangnam-gu,
Seoul, 06241, Republic of Korea www.lunit.io

Page 3/6

Summary of Substantial Equivalence ട.

	Subject Device	Predicate Device
Item	Lunit INSIGHT DBT	Lunit INSIGHT MMG
Classification Name	Radiological Computer AssistedDetection/Diagnosis Software For SuspiciousLesions For Cancer	Radiological Computer AssistedDetection/Diagnosis Software For SuspiciousLesions For Cancer
Regulation	21 CFR 892.2090	21 CFR 892.2090
Regulatory Class	Class II	Class II
Product Code	QDQ	QDQ
Indication for Use	Lunit INSIGHT DBT is a computer-assisteddetection and diagnosis (CADe/x) softwareintended to be used concurrently byinterpreting physicians to aid in the detectionand characterization of suspected lesions forbreast cancer in digital breast tomosynthesis(DBT) exams from compatible DBT systems.Through the analysis, the regions of softtissue lesions and calcifications are markedwith an abnormality score indicating thelikelihood of the presence of malignancy foreach lesion. Lunit INSIGHT DBT uses screeningmammograms of the female population.Lunit INSIGHT DBT is not intended as areplacement for a complete interpretingphysician's review or their clinical judgmentthat takes into account other relevantinformation from the image or patienthistory.	Lunit INSIGHT MMG is a radiologicalComputer-Assisted Detection and Diagnosis(CADe/x) software device based on anartificial intelligence algorithm intended to aidin the detection, localization, andcharacterization of suspicious areas for breastcancer on mammograms from compatibleFFDM systems. As an adjunctive tool, thedevice is intended to be viewed byinterpreting physicians after completing theirinitial read. It is not intended as areplacement for a complete physician'sreview or their clinical judgement that takesinto account other relevant information fromthe image or patient history. The LunitINSIGHT MMG uses screening mammogramsof the female population.
Target patientpopulation	Women undergoing mammography	Women undergoing mammography
Intended user	Physicians interpreting screeningmammograms	Physicians interpreting screeningmammograms
Input Image Source	DBT	FFDM
FundamentalTechnological Basis	Lunit INSIGHT DBT is powered by artificialintelligence/machine learning-based softwarealgorithm	Lunit INSIGHT MMG is powered by artificialintelligence/machine learning-based softwarealgorithm

{6}------------------------------------------------

Image /page/6/Picture/0 description: The image shows the Lunit logo. The logo consists of a blue circle with a white molecular-like structure inside, followed by the word "Lunit" in black font. A registered trademark symbol is located to the right of the word "Lunit".

am-daero. Gangnam-gu.

Page 4/6

6. Comparison with Predicate Device

The substantial equivalence table above summarizes the similarities and differences between Lunit INSIGHT DBT and its predicate device, Lunit INSIGHT MMG (K211678). Both devices use artificial intelligence technologies and deep learning techniques to fulfill its intended purpose to detect and characterize lesions suspected of breast cancer. The devices differ in its input file for analysis where Lunit INSIGHT DBT requires its predicate analyzes FFDM's. Outputs of both devices augments the interpreting physicians in the diagnosis of asymptomatic patients.

7. Performance Data

7.1. Non-clinical Testing Summary

Software Verification and Validation

Lunit INSIGHT DBT is determined as Moderate level of Concern since a malfunction of, or a latent design flaw in, the software could result in Minor injury. Software was verified through software integration test and software system test. Based on results of verification, Lunit INSIGHT DBT demonstrated that it fulfilled the software requirements.

Standalone Performance Testing

A standalone performance study of the Lunit INSIGHT DBT assessed the detection performance of the artificial intelligence algorithm for breast cancer within DBT exams.

Total of 2,202 DBT exams of female adults were collected at multiple imaging facilities in the US using Hologic and GE Healthcare equipment. The data was collected consecutively with the following information: patient information, original radiology report, follow-up biopsy and pathology data, and further imaging diagnostic workup. The dataset consisted of 1,100 negative and benign cases, and 1,102 cancer cases. In terms of ethnicity and race, the cases were composed of White, American Indian, African, Asian, and other races, and representative of the general US population. The standalone performance of the Lunit INSIGHT DBT was examined by comparing the analysis results with the reference standards. The reference standards were established through binary classification of each case based on clinical supporting data, particularly pathology reports for cancer and biopsy-proven benign cases, followed by localization which was derived based on the radiologic review and annotation by multiple MQSA qualified ground truthers. The dataset used in the standalone performance test was independent from the dataset used for development of the artificial intelligence algorithm. For generalizability, various subgroup analyses were conducted on the collected dataset including image/radiologic characteristics (e.g. modality manufacturer, slice thickness), demographic information (e.g., age, race), and clinically relevant confounders (e.g. breast cancer type),

{7}------------------------------------------------

Image /page/7/Picture/0 description: The image shows the logo for Lunit. The logo consists of a blue circle with a white molecular-like structure inside, followed by the word "Lunit" in black, sans-serif font. A registered trademark symbol is placed to the upper right of the word "Lunit".

am-daero Gangnam-gu

Page 5/6

Standalone Performance Results

The primary endpoint was to demonstrate AUROC in standalone performance greater than 0.903, the mean AUROC of the predicate device (K211678). The subject device's AUROC in the standalone performance analysis was 0.928 (95% Cl: 0.917 - 0.939) with statistical significance (p < 0.0001), which exceeded the acceptance criteria of the primary endpoint.

7.2. Clinical Assessment Summary

Clinical performance assessment was conducted to evaluate effectiveness of Lunit INSIGHT DBT in the assistance of detection and diagnosis of breast cancer during DBT exam interpretation. A retrospective, multi-reader multicase (MRMC) study was conducted comparing the reading panel's interpretation performance with and without the use of the Lunit INSIGHT DBT software during the DBT exam interpretation. During the study, every reading panel member, a total of 15 MQSA qualified and US board-certified radiologists, performed interpretation and completed reading sessions, CAD unassisted and CAD assisted, independently using a setting similar to a screening procedure in the US.

Clinical Assessment Primary Objective

The primary objective of the clinical performance assessment was to evaluate the effectiveness of Lunit INSIGHT DBT by comparing the clinical performance of radiologists with CAD and without CAD assistance. If the performance with CAD assistance is superior to that of without CAD assistance with statistical significance, the study was considered to be successful.

Clinical Assessment Data Description

Total of 258 DBT exams were acquired from US clinical centers and were collected using Hologic and GE Healthcare equipment. 65 were cancer cases and 193 were non-cancer cases (128 normal and 65 benign cases).

Clinical Assessment Results

The primary endpoint result of the study was comparison of patient-level Level of Suspicion (LOS) area under the Receiver Operating Characteristic (ROC) curve between CAD-assisted interpretation. AUROC for CAD-unassisted interpretation was 0.897 (95% Cl 0.858 - 0.936), when that of CAD-assisted interpretation was 0.915 (95% Cl: 0.874 - 0.955) with inter-test difference of 0.017 (95%: Cl 0.000 - 0.034, P = 0.0498).

8. Assessment of Benefit-Risk, General Safety and Effectiveness

Risk management of the subject device is conducted via hazard analysis which identifies and mitigates existing and potential hazards. Hazards were controlled throughout the software lifecycle with control measures with regards to software development, verification, and validation. Furthermore, labeling information consists of instructions for use with necessary cautionary statements for safe and effective use of the software. Lunit finds the use of the software has a positive balance in terms of probable benefits versus foreseeable and identified risks.

{8}------------------------------------------------

Image /page/8/Picture/0 description: The image shows the Lunit logo. The logo consists of a blue circle with a white molecular structure inside, followed by the word "Lunit" in black, with a registered trademark symbol next to it. The logo is simple and modern, and the colors are eye-catching.

Lunit Inc. 4-8 F, 374, Gangnam-daero, Gangnam-gu, Seoul, 06241, Republic of Korea www.lunit.io

Page 6/6

9. Conclusion

Lunit INSIGHT DBT is substantially equivalent to Lunit INSIGHT MMG because they are identical with regards to intended use and share similar technological or performance characteristics. The minor differences in technological characteristics do not alter the intended use of the device and do not raise new questions or safety and effectiveness. In addition, non-clinical and clinical testing results demonstrate that the Lunit INSIGHT DBT is as safe and effective as the predicate Lunit INSIGHT MMG. Thus, the substantial equivalence has been demonstrated.

Regulation Number and Section

§ 892.2090 Radiological computer-assisted detection and diagnosis software.

(a)
Identification. A radiological computer-assisted detection and diagnostic software is an image processing device intended to aid in the detection, localization, and characterization of fracture, lesions, or other disease-specific findings on acquired medical images (e.g., radiography, magnetic resonance, computed tomography). The device detects, identifies, and characterizes findings based on features or information extracted from images, and provides information about the presence, location, and characteristics of the findings to the user. The analysis is intended to inform the primary diagnostic and patient management decisions that are made by the clinical user. The device is not intended as a replacement for a complete clinician's review or their clinical judgment that takes into account other relevant information from the image or patient history.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Design verification and validation must include:
(i) A detailed description of the image analysis algorithm, including a description of the algorithm inputs and outputs, each major component or block, how the algorithm and output affects or relates to clinical practice or patient care, and any algorithm limitations.
(ii) A detailed description of pre-specified performance testing protocols and dataset(s) used to assess whether the device will provide improved assisted-read detection and diagnostic performance as intended in the indicated user population(s), and to characterize the standalone device performance for labeling. Performance testing includes standalone test(s), side-by-side comparison(s), and/or a reader study, as applicable.
(iii) Results from standalone performance testing used to characterize the independent performance of the device separate from aided user performance. The performance assessment must be based on appropriate diagnostic accuracy measures (
e.g., receiver operator characteristic plot, sensitivity, specificity, positive and negative predictive values, and diagnostic likelihood ratio). Devices with localization output must include localization accuracy testing as a component of standalone testing. The test dataset must be representative of the typical patient population with enrichment made only to ensure that the test dataset contains a sufficient number of cases from important cohorts (e.g., subsets defined by clinically relevant confounders, effect modifiers, concomitant disease, and subsets defined by image acquisition characteristics) such that the performance estimates and confidence intervals of the device for these individual subsets can be characterized for the intended use population and imaging equipment.(iv) Results from performance testing that demonstrate that the device provides improved assisted-read detection and/or diagnostic performance as intended in the indicated user population(s) when used in accordance with the instructions for use. The reader population must be comprised of the intended user population in terms of clinical training, certification, and years of experience. The performance assessment must be based on appropriate diagnostic accuracy measures (
e.g., receiver operator characteristic plot, sensitivity, specificity, positive and negative predictive values, and diagnostic likelihood ratio). Test datasets must meet the requirements described in paragraph (b)(1)(iii) of this section.(v) Appropriate software documentation, including device hazard analysis, software requirements specification document, software design specification document, traceability analysis, system level test protocol, pass/fail criteria, testing results, and cybersecurity measures.
(2) Labeling must include the following:
(i) A detailed description of the patient population for which the device is indicated for use.
(ii) A detailed description of the device instructions for use, including the intended reading protocol and how the user should interpret the device output.
(iii) A detailed description of the intended user, and any user training materials or programs that address appropriate reading protocols for the device, to ensure that the end user is fully aware of how to interpret and apply the device output.
(iv) A detailed description of the device inputs and outputs.
(v) A detailed description of compatible imaging hardware and imaging protocols.
(vi) Warnings, precautions, and limitations must include situations in which the device may fail or may not operate at its expected performance level (
e.g., poor image quality or for certain subpopulations), as applicable.(vii) A detailed summary of the performance testing, including test methods, dataset characteristics, results, and a summary of sub-analyses on case distributions stratified by relevant confounders, such as anatomical characteristics, patient demographics and medical history, user experience, and imaging equipment.